Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
jakewins
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
72 ms
·
421.
▲
by
jakewins
8y ago
So the US averages a ratio of 10:1 administrative employees[0] per doctor. As a random comparison, Sweden maintains roughly a 1:1 ratio [1]. Ironically, the link below is a Swedish article about how the Swedish admin-to-doctor ratio is bloa
422.
▲
by
jakewins
8y ago
Here are transcripts and documents from the case as its progressed of anyone is interested in how the jury was convinced: https://www.baumhedlundlaw.com/toxic-tort-law/monsanto-round...
423.
▲
VC to pay $20M damages to founders for “destroying [Riaks] value entirely”
(law360.com)
2 points
by
jakewins
8y ago
|
0 comments
424.
▲
by
jakewins
8y ago
Yes - the cache example is bad; my point is the general idea of avoiding branches that merge back into the main function. In your example above, the branch does merge back, and clearly that's fine for a small function - and common patt
425.
▲
by
jakewins
8y ago
Another pattern I've found helpful is to have branches "exit early"; if you have code like: if !cache.contains(key) { v = db.query("..") expiry = calcexpiry(v) cache.put(k, v, expiry) r
426.
▲
by
jakewins
8y ago
Same as how TV and Radio ratings are done - statistically significant sampling of the population via surveys: https://support.alexa.com/hc/en-us/articles/200449744-How-ar...
427.
▲
A solution to byzantine consensus without proof-of-X [PDF]
(drive.google.com)
2 points
by
jakewins
8y ago
|
1 comments
428.
▲
by
jakewins
8y ago
> they have sued over seed saving, such as the story of Michael White. Here is the case with Mr. White: https://www.gpo.gov/fdsys/pkg/USCOURTS-alnd-5_03-cv-02804/pd... As you can see by reading it, it is n
429.
▲
by
jakewins
8y ago
It seems this number is for soil decomposition from microbial action. It's stable to chemical and photo decomposition. It also seems I was wrong about half-life in plants; the half life given for some tree species is 9 days. Unsurprisi
430.
▲
by
jakewins
8y ago
Not to be a Monsanto shill, but: > [forces patent compliance] even in situations where it is clear Monsanto crops contaminated someone else's field Can you provide any example of this? The only cases I'm aware of are the likes
431.
▲
by
jakewins
8y ago
I think you are confusing "degrade" with "run-off"; glyphosate "sticks" to soils with high organic content - presumably due to opposing electric charges. The point of that being that it doesn't get washed
432.
▲
by
jakewins
9y ago
A permanent function that tries to break into my systems in order to create structural pressure to think about security. https://en.wikipedia.org/wiki/Red_team
433.
▲
Ask HN: Is there a “red-team-as-a-service” you recommend?
1 points
by
jakewins
9y ago
|
2 comments
434.
▲
by
jakewins
9y ago
Ugh, this guy calls me several times a week. I'm on the no-call list, I report them to the FTC every time they call, and twice I've waited to get patched through to the "travel rep" to ask them to remove me. It's in
435.
▲
by
jakewins
9y ago
On the one hand, I'm rather happy about this sudden outpouring of anger; FB has fundamentally corrupted the public square, scrutiny is appropriate. On the other hand, it's hard not to get conspiratorial, given https://w
436.
▲
by
jakewins
9y ago
> GDPR is ripe for the same treatment Nonsense. Patent litigation allows trolls to sue for and receive damages from the other party. GDPR violations lead to fines fines that go to the government ; the plaintiff does not receive damages.
437.
▲
by
jakewins
9y ago
I don't think this is correct, do you have a source for this? man PTHREAD_ATTR_SETSTACKSIZE(3) says: > The stack size attribute determines the minimum size (in bytes) that will be allocated for threads created using the thread attri
438.
▲
by
jakewins
9y ago
This is totally true, and I agree; I tried to guard against this at the end with the "I think this is enough to state a hypothesis , which would then need to be tested.." paragraph. However - the same is true for 2009, 2010, 2011
439.
▲
by
jakewins
9y ago
> Don't tell us how safe Tesla's are compared to the entire US driving population: Tell us how safe they are compared to equivalent vehicles from other manufacturers. I'm having a hard time to find an exact number comparis
440.
▲
by
jakewins
9y ago
Stacks in Linux grow dynamically, so if you want go-like stacks you just set the max stack size == system memory. See http://man7.org/linux/man-pages/man2/setrlimit.2.html
441.
▲
by
jakewins
9y ago
> They track you and keep a profile on you no matter what. This is only true because the voting US public believes it to be true. Meanwhile, the EU simply wrote a law against it. Now any EU citizen can simply withhold consent, and if FB
442.
▲
by
jakewins
9y ago
Yeah.. I'm confused; I return a lot of stuff at hardware stores, because I suck at buying the right things. If I don't have a receipt, returns differ by brand: - Menards; just added ability to look up your receipt yourself at thei
443.
▲
by
jakewins
9y ago
"Exploitation requires that an attacker be able to run a program with local-machine elevated administrator privileges." https://safefirmware.com/amdflaws_whitepaper.pdf I wish Intel would invest in fixing their CP
444.
▲
by
jakewins
9y ago
Stevia does produce an insulin response - but, it appears, in the "right" direction. There are apparently several labs exploring stevia as a treatment in diabetes models, for instance: https://www.thieme-connect.com
445.
▲
by
jakewins
9y ago
> on't have any effect on your blood glucose and really are completely different Source for this? My understanding is that the link between non-caloric artificial sweeteners (NAS) and diabetes is growing increasingly clear. For inst
446.
▲
by
jakewins
9y ago
On (1), the memory layer has been entirely rewritten since 2011. It used to be a combination of MMAP and on-heap caching; mmap in java being notoriously terrible and caching on the heap being even worse. The memory layer now works similar t
447.
▲
by
jakewins
9y ago
Pumps are not good at stopping and starting. They like to start, run for at least a few minutes, and then stop. Each stop/start cycle adds wear to the pump, as the components inside it are at peak strain and friction when they speed up
448.
▲
by
jakewins
9y ago
Hah :) I don't mean that malloc itself is fast, I mean that having non-jvm heap memory is fast. A permanent memory block on the JVM heap can't use pointers to refer to it, since GC moves objects around. And even though those block
449.
▲
by
jakewins
9y ago
This is correct; the standard approach here is to use regular c-style memory management for the data the system is managing, and the JVM heap only for the database "infrastructure". This hybrid approach gives the benefit of a mana
450.
▲
by
jakewins
9y ago
> If the vehicle is a fixed cost for the person regardless of whether or not they are a driver What? Vehicles are clearly not fixed cost, they are assets that deprecate in value from use. There is some deprecation from age, but by far
More ›